Start with what "the market decided" actually hides, because political power hides the same way. This site has already made the economic version of this argument: farm subsidies and coal bailouts get described as neutral outcomes when they're actually a named vote, on a specific date, by a specific body. Empires work the same way at the largest scale -- Egypt, Babylon, Rome, and Ming China all rose and fell on a real, measurable cycle, and every turn of that cycle was still somebody's specific act of will on a specific day. Political momentum inside a democracy runs on the identical mechanism, just faster and more visible: "Sanders has momentum in the Northeast" sounds like a weather pattern. It's actually one person, showing up, in a specific place, a specific number of times, and the pattern only becomes visible once you look at more than one election.

New Hampshire, twice. Bernie Sanders won the 2016 New Hampshire Democratic primary by 22 points -- 60% to Hillary Clinton's 38%.^[1] Four years later, in a much more crowded field, he won again, narrowly, with 26% to Pete Buttigieg's nearly 25% and Amy Klobuchar's 20%.^[2] New Hampshire borders Vermont, and Sanders had been Vermont's own senator since 2007 -- not a national figure parachuting in every four years, but someone whose name New Hampshire voters had had a decade to actually know before he first appeared on their primary ballot.^[2] That's the same underlying mechanism as the door-by-door piece already on this site -- Mamdani in Astoria, El-Sayed's volunteers in Michigan -- proximity substituting for money and media. New Hampshire didn't just have a retail-politics culture in the abstract. It had a decade of actual proximity to one specific candidate before he ran.

Michigan, three times, including the time it failed. In 2016, Sanders won Michigan's Democratic primary in one of the largest polling misses in three decades -- polling averages had Clinton ahead by 21 points; Sanders won 49.8% to 48.3%, a margin nobody's model saw coming, driven in large part by youth turnout pollsters undercounted.^[3] In 2018, Sanders returned to Michigan in person, campaigning alongside Alexandria Ocasio-Cortez -- fresh off her own primary upset in New York weeks earlier -- for Abdul El-Sayed's run for governor.^[4] El-Sayed lost that race to Gretchen Whitmer. Worth stating plainly rather than skipping past: the lever doesn't move every time it's pulled. But the same El-Sayed, running for US Senate in 2026, beat more than $60 million in outside spending with roughly 12,000 door-knocking volunteers and won by under one point -- a result already documented on this site.^[5] Three elections, one state, one recurring name in the middle of all of them, one real loss along the way.

The pattern isn't "Sanders is popular." It's that a specific act, repeated, is traceable -- and a weather-system description of politics makes it untraceable on purpose. Calling a result "momentum" or "an energized base" describes an outcome without naming who did what, when, to produce it. Naming the actual sequence -- a decade of border-state proximity before 2016, a return trip in 2018 that didn't work, the same network's volunteer-first method winning in 2026 -- turns an outcome that reads as inevitable into a specific, buildable, repeatable mechanism, with a documented failure inside it. That's the same distinction "the trade we keep remaking" draws for economic policy: 1790's debt assumption and 2020's airline bailout both look like neutral, structural events until you name the specific rooms and specific trades that actually produced them. Political power and economic policy get described impersonally for the same reason -- naming who actually moved the lever also names who's accountable for it, win or lose.